presentation of the coherent project iwg-dpps

TU Graz I Vehicle Safety Institute

www.tugraz.at

1

S C I E N C E P A S S I O N T E C H N O L O G Y

u www.tugraz.at

Presentation of the CoHerent ProjectIWG-DPPSCorina Klug, Florian Feist, Wolfgang Sinz, James Ellway, Michiel van Ratingen

Source: Corina Klug, Florian Feist, Wolfgang Sinz, James Ellway, Michiel van Ratingen: „A Procedure to compare kinematics of Human Body Models for pedestrian assessments”, presented at SAE Government/Industry Meeting January 25th, 2018

19.04.2018

IWG-DPPS/1/04


www.tugraz.at

2

• Improving safety assessment with HBMs:

• Influence of body size

• Addressing multiple scenarios

• Additional injury predictors

Motivation – Application of Human Body Models

ENHANCE PEDESTRIAN SAFETY WITH IMPROVED ASSESMENT METHOD

Biomechanical Research

Product Development

Standards and NCAPs

http://www.elemance.com

CoHerent 19.04.2018

IWG-DPPS/1/04


www.tugraz.at

3

Euro NCAP Assessment of Deployable Systems - A Hybrid Approach

RatingSimulation with HBMs Sub-system Tests

Head Injury Criterion

• Head Contact Time• Head Impact Location• Deflection due to body

loading

COMPARABILITY!!!

CoHerent 19.04.2018

IWG-DPPS/1/04


www.tugraz.at

4

• Variety of HBMs based on different anthropometry and validated with different PMHS tests

• Variety of Versions of HBMs (in-house modifications & revisions)

• Variety of Solvers

• Variety of Initial Positions

• Variety of Simulation Settings

Challenges

Source: ACEA, 2016

CoHerent 19.04.2018

IWG-DPPS/1/04


www.tugraz.at

5

• AIM: Harmonization of Pedestrian Simulations with Human Body Models

• How are kinematics affected by varying simulation setups?

• How are kinematics affected by varying pedestrian models?

• How to ensure that pedestrian assessment simulations with different HBMs render consistent results?

Objectives

Procedure for Kinematic Comparison of HBMs needed, applicable for • varying HBMs in • varying codes

CoHerent 19.04.2018

IWG-DPPS/1/04


www.tugraz.at

6

Comparison of HBMs

Method

Generic Vehicle Models

• Representative• Transferrable to

other solvers

Harmonised Simulation Setup

Simulation Protocol

Certification Procedure for Euro NCAP

Sensitivity Study

Two HBMs in one code

THUMS v 4.02

GHBMC PS v1.4.3

AM 50

Application on more models in 4

codes

Harmonised Postprocessing

Corridors

136

138

140

142

HIT

[m

s]

19.04.2018CoHerent

IWG-DPPS/1/04


www.tugraz.at

7

Comparison of HBMs

Method



other solvers


Simulation Protocol


Sensitivity Study


THUMS v 4.02

GHBMC PS v1.4.3

AM 50


codes


Corridors

136

138

140

142

HIT

[m

s]

19.04.2018CoHerent

IWG-DPPS/1/04


www.tugraz.at

8


0

10

0 20 40 60 80

Forc

e [k

N]

Corridor Loading

Corridor Unloading

Median Loading

Median Unloading

Generic Vehicle Model

Generic Vehicle Model

0

5

10

15

0 20 40 60 80

Forc

e [k

N]

Deflection [mm]

SUV

MPV

Familycar

Roadster

CoHerent 19.04.2018

IWG-DPPS/1/04


www.tugraz.at

9

Comparison of HBMs

Method



other solvers


Simulation Protocol


Sensitivity Study


THUMS v 4.02

GHBMC PS v1.4.3

AM 50


codes


Corridors

136

138

140

142

HIT

[m

s]

19.04.2018CoHerent

IWG-DPPS/1/04


www.tugraz.at

10

• effect of arm posture: up to 4% in HIT [1]

• effect of contact setting: up to 3% in HIT [1]

Sensitivity Study

Proper protocol for virtual testing needed

136138140142144

Pos1 Pos2 Pos3

HIT

[m

s]

GHBMC Inital Positions

136138140142

HIT

[m

s]

conatct variation

CoHerent 19.04.2018

IWG-DPPS/1/04


www.tugraz.at

11

Comparison of HBMs

Method



other solvers


Simulation Protocol


Sensitivity Study


THUMS v 4.02

GHBMC PS v1.4.3

AM 50


codes


Corridors

136

138

140

142

HIT

[m

s]

19.04.2018CoHerent

IWG-DPPS/1/04


www.tugraz.at

12

• THUMS v4.02 and GHBMC PS v1.4.3 in LS-DYNA show very comparable kinematics

• Difference in HIT for “Family car” impact at 40 kph = 0.7 ms

Comparison of two HBMs

19.04.2018CoHerent

IWG-DPPS/1/04


www.tugraz.at

13

-3%

-2%

-1%

0%

1%

2%

3%

30kph 40kph 50kph 30kph 40kph 50kph 30kph 40kph 50kph 30kph 40kph 50kph

FCR MPV RDS SUV

Difference in HIT values between THUMS v4 and GHBMC PS

Comparison of two HBMs – Difference in Head Impact Time

CoHerent 19.04.2018

IWG-DPPS/1/04


www.tugraz.at

14

Max. difference of location of head COG in “Roadster” impact at 30 kph – 4% of path length

Comparison of two HBMs – Maximum Difference in Trajectories

GHBMCTHUMS

HIT = 180.2HIT =180.8

CoHerent 19.04.2018

IWG-DPPS/1/04


www.tugraz.at

15

Harmonized protocol was applied with varying models in multiple codes on different platforms at several institutions

Application to Various HBMs in Multiple Codes

HIT

Model

80

85

90

95

100

105

110

1 2 3 4 5 6 7 8 9 101112131415161718

HCT

all Models

THUMS v4and GHBMC

0

500

1000

1500

2000

2500

3000

-2500 -1500 -500 500

MPV 50 kph

0

500

1000

1500

2000

2500

3000

-2500 -1500 -500 500

Head trajectoryMPV 50 kph

CoHerent 19.04.2018

IWG-DPPS/1/04


www.tugraz.at

16

Comparison of HBMs

Method



other solvers


Simulation Protocol


Sensitivity Study


THUMS v 4.02

GHBMC PS v1.4.3

AM 50


codes


Corridors

136

138

140

142

HIT

[m

s]

19.04.2018CoHerent

IWG-DPPS/1/04


www.tugraz.at

17

0

500

1000

1500

2000

2500

3000

-2500 -2000 -1500 -1000 -500 0 500

Head trajectory

• Results from all parties were collated and analysed

• Outliers were identified and eliminated - corridors were created from „consistent“ results

• Trajectory is considered time dependent as timing is essential for evaluation

Certification Procedure for Euro NCAP - Trajectories

outliers

-1000

-800

-600

-400

-200

0

200

400

0 100 200

x-Tr

ajec

tory

[m

m]

Time [ms]

0

500

1000

1500

2000

0 100 200

z-Tr

ajec

tory

[m

m]

Time [ms]

zmax (tn)

zmin (tn)

xmin (tn) xmax (tn)

tn-1

tn

z(x)0

500

1000

1500

2000

0 100 200

z-Tr

ajec

tory

[m

m]

Time [ms]

-1000

-800

-600

-400

-200

0

200

400

0 100 200

x-Tr

ajec

tory

[m

m]

Time [ms]

zmax (tn)

zmin (tn)

xmin (tn) xmax (tn)

tn-1

tn

z(x)

loca

l z c

oo

rdin

ate

[mm

]

local x coordinate [mm]

AC

T12

HC

0

500

1000

1500

2000

2500

3000

-2500 -2000 -1500 -1000 -500 0 500

loca

l z c

oo

rdin

ate

[mm

]


AC

T12

HC

0

500

1000

1500

2000

2500

3000

-2500 -2000 -1500 -1000 -500 0 500

loca

l z c

oo

rdin

ate

[mm

]


CoHerent 19.04.2018

IWG-DPPS/1/04


www.tugraz.at

18

outliers

-20%-15%-10%

-5%0%5%

10%15%20%


FCR MPV RDS SUV

HIT

-20%-15%-10%

-5%0%5%

10%15%20%


FCR MPV RDS SUV

HIT

• Results from all parties were collated and analysed

• Outliers were identified and eliminated - corridors were created from „consistent“ results

Certification Procedure for Euro NCAP – Head Impact Time

-20%

-15%

-10%

-5%

0%

5%

10%

15%

20%


FCR MPV RDS SUV

HIT

+3.5%

- 7%

Median of all consistent results

CoHerent 19.04.2018

IWG-DPPS/1/04


www.tugraz.at

19

Certification Report

19.04.2018CoHerent

FCR 40 kph

Pedestrian Model submitted by Date

e.g. THUMS Manufacture XY 2017-10-10 check value

OK

OK visual check

OK visual check

OK 0

OK 0.0%

OK 0.0%

OK 0.0%

OK 0.0%

OK 0.10%

OK max. 0mm 0ms

OK -0.9% (HIT=136.9ms)

max. 0kN 0ms

t_min= 4E-04

Artificial mass increase for moving parts <= 3%

All required data provided?

FE Surfaces getting in contact do not cross each other

Surfaces getting in contact do not get trapped one in the other (sticky nodes)

Contact force (between HBM and vehicle) is zero at simulation start

Total energy remains constant within a 15% tolerance

Hourglass energy <= 10% of the total energy

Contact energy at the simulation start <= 1% of the total energy

Artificial energy (contact energy and hourglass energy) <= 15% of the total energy

Trajectories are within corridor

HIT is within tolerance

Contact Force is within corridor (monitoring only!)

Time step does not fall and stays excessively low (monitoring only)

Templates are available on Euro NCAP website as

„Certification Pack“https://www.euroncap.com/en/for-engineers/supporting-information/

IWG-DPPS/1/04

https://www.euroncap.com/en/for-engineers/supporting-information


www.tugraz.at

20

Check Generic Vehicle Models

Certification of HBM

Assessment of Deployable System

With consistent settings and unchanged HBM

Euro NCAP Procedure for Virtual Pedestrian Safety Assessment

HBM vs. full FE vehicle model

HBM vs. GV Models4 vehicle shapes, 30, 40 and 50 km/h

Impactor vs. GV Models

Pedestrian models and setup has to be certified according to TB 024 before being used in deployable bonnet simulations

CoHerent 19.04.2018

[2]

IWG-DPPS/1/04


www.tugraz.at

21

• Certification procedure does not replace the validation of HBMs – it qualifies model and setup for safety assessment to allow HBM to be used as a “virtual test device”

• Although GHBMC and THUMS are validated with differing PMHS tests, and do not have the exact same geometry, the response was very similar - This wasn’t true for some other pedestrian models

• Only kinematics were compared, no injury metrics

• Contact force corridors were derived for guidance only

• Corridors currently only available for 50th percentile male

Discussion and Limitations

19.04.2018CoHerent

IWG-DPPS/1/04


www.tugraz.at

22

• Inconsistent boundary conditions can lead to larger differences in terms of HIT than differences observed between the two HBMs in one code

• When setup was harmonized, THUMS v4.02 and GHBMC v1.4.3 led to very comparable results – an important outcome for the Euro NCAP assessment

• Application of HBMs for pedestrian assessment is not straightforward - challenges have to be addressed

• Procedure developed to ensure consistency between Human Body Models has been adopted by Euro NCAP as of Jan 2018

Conclusions

19.04.2018CoHerent

IWG-DPPS/1/04


www.tugraz.at

23

• The project CoHerent was funded by Euro NCAP.

• The authors would like to thank ACEA for provision of data for GV models and financial support of the project.

• We acknowledge that Elemance LLC is the exclusive distributor of the Global Human Body Model Consortium-owned GHBMC-Model.

• We acknowledge that Toyota Motor Corporation and Toyota Central R&D labs, Inc., are the owners of the licensed Total Human Body Model for Safety.

• The authors would like to acknowledge the use of HPC resources provided by the ZID of Graz University of Technology.

• We would like to thank all participants of the CoHerent project for their contribution to the project (TASS, SUBARU, Nissan, JLR, HongIK University, Daimler, BMW, JARI, WFU, Audi, Honda, ESI, Simulia, Dassault Systemes, Altair).

Acknowledgement

CoHerent 19.04.2018

IWG-DPPS/1/04


www.tugraz.at

24

Corina Klug

Graz University of Technology

Inffeldgasse 23\I, 8010 Graz, Austria

+43 316 873 30329

[email protected]

References: 1. Klug C, Feist F, Raffler M, Sinz W, Ellway J, Petit P, van Ratingen, M. “Development of a Procedure to Compare Kinematics of Human Body Models for

Pedestrian Simulations”. In: 2017 IRCOBI Conference Proceedings; 2017.

2. Euro NCAP. “TB024 - Pedestrian HBM Certification” https://cdn.euroncap.com/media/34544/tb-024-pedestrian-human-model-certification-v101.pdf

3. Klug C, Feist F, Raffler M, Sinz W, Ellway J, van Ratingen, M. : „A Procedure to compare kinematics of Human Body Models for pedestrian assessments”, presented at SAE Government/Industry Meeting, January 25th, 2018

Contact

CoHerent 19.04.2018

IWG-DPPS/1/04

mailto:[email protected]

https://cdn.euroncap.com/media/34544/tb-024-pedestrian-human-model-certification-v101.pdf

presentation of the coherent project iwg-dpps

Documents